Download Knowledge Helps: Mechanistic Information and Numeric Evidence as Cognitive

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Myron Ebell wikipedia , lookup

Mitigation of global warming in Australia wikipedia , lookup

2009 United Nations Climate Change Conference wikipedia , lookup

Michael E. Mann wikipedia , lookup

Soon and Baliunas controversy wikipedia , lookup

Climate resilience wikipedia , lookup

Climatic Research Unit email controversy wikipedia , lookup

Effects of global warming on human health wikipedia , lookup

General circulation model wikipedia , lookup

ExxonMobil climate change controversy wikipedia , lookup

Heaven and Earth (book) wikipedia , lookup

Economics of global warming wikipedia , lookup

Climate sensitivity wikipedia , lookup

Climate engineering wikipedia , lookup

Citizens' Climate Lobby wikipedia , lookup

Instrumental temperature record wikipedia , lookup

Fred Singer wikipedia , lookup

Global warming controversy wikipedia , lookup

Climate governance wikipedia , lookup

Global warming hiatus wikipedia , lookup

Climate change adaptation wikipedia , lookup

Climatic Research Unit documents wikipedia , lookup

Climate change denial wikipedia , lookup

Carbon Pollution Reduction Scheme wikipedia , lookup

United Nations Framework Convention on Climate Change wikipedia , lookup

Global warming wikipedia , lookup

Global Energy and Water Cycle Experiment wikipedia , lookup

Climate change in Tuvalu wikipedia , lookup

Climate change and agriculture wikipedia , lookup

Effects of global warming wikipedia , lookup

Politics of global warming wikipedia , lookup

Climate change in the United States wikipedia , lookup

Solar radiation management wikipedia , lookup

Attribution of recent climate change wikipedia , lookup

Effects of global warming on humans wikipedia , lookup

Climate change and poverty wikipedia , lookup

Climate change feedback wikipedia , lookup

Media coverage of global warming wikipedia , lookup

Scientific opinion on climate change wikipedia , lookup

Climate change, industry and society wikipedia , lookup

Public opinion on global warming wikipedia , lookup

IPCC Fourth Assessment Report wikipedia , lookup

Surveys of scientists' views on climate change wikipedia , lookup

Transcript
Knowledge Helps: Mechanistic Information and Numeric Evidence as Cognitive
Levers to Overcome Stasis and Build Public Consensus on Climate Change
1
2
Dav Clark ([email protected]) 1
Michael Andrew Ranney ([email protected]) 1,2
Jacqueline Felipe ([email protected]) 2
Department of Psychology, 3210 Tolman Hall; Berkeley, CA 94720-1650 USA
Graduate School of Education, 4533 Tolman Hall; Berkeley, CA 94720-1670 USA
A clear, growing consensus indicates an urgent need for
humans to reduce the burgeoning effects of global climate
change (“global warming” or GW). Apt public instruction
seems central to achieving critical behavioral changes, but
some researchers suggest that U.S. climate attitudes are
doomed to cognitive stasis (i.e., that little will be gained by
educating the public). Herein are four studies that counter the
stasis view. Our laboratory has previously reported findings
that (1) virtually no Americans know the basic climate change
mechanism, yet it (2) is quickly learned (in a few minutes,
e.g., with a 400-word text), which (3) increases climate
change acceptance. Below, Studies 1 and 4 replicate and
extend these results to demonstrate (a) efficacy with an online
presentation and broader populations and (b) retention up to a
month after learning the mechanism. Studies 2-4 explore roles
for germane numerical information using estimation with
feedback. Study 2 shows that (d) misleading, cherry-picked,
statistics can decrease climate change acceptance (and shake
metacognition), while Studies 3 and 4 show that (e) surprising
scientific information must be presented with care for it to
foster beliefs in line with climate science’s consensus. In sum,
contrary to unnecessarily pessimistic (and correlational)
"stasis" arguments, highly germane science information can
clearly change the public’s understandings and opinions.
Keywords: cognitive change, science education, explanation,
climate change, global warming, acceptance, mechanism.
Climate Change as a Behavioral Problem
Our atmosphere’s carbon dioxide (CO2) concentration is
higher now than in any of the past 15 million years (World
Bank, 2012). Global warming (GW) akin to recent trends
last occurred over 17 million years ago, when a 3-4°C gain
occurred over 1,500,000 years. Standard models show that
continuing our current behavior will yield similar warming
in just 100 years. In previous warming periods of this
magnitude, widespread extinctions occurred. With imminent
warming 10,000+ times faster than historical timescales, the
biological systems we depend upon (e.g., for food) will
clearly be severely impacted (Barnosky, 2009). Nearly all
climate researchers have concluded that the problem is
urgent and anthropogenic (i.e., essentially 100% humancaused). It is thus behavioral, and will be "solved" only by
changes in human behavior. The IPCC (Intergovernmental
Panel on Climate Change) and Skeptical Science have
assembled and disseminated the scientific consensus on
GW, but, sadly, the U.S. public is still divided on both
GW’s existence and its cause (cf. Hoffman, 2011).
A group of climate communication researchers, oddly,
suggests that educational ventures would be of little or no
help. Kahan et al. (2012) found (through correlational
means) that, for the U.S. (a high per-capita carbon user),
direct cognitive approaches (including numeracy and
science education) seem to solidify biased views––
reinforcing a kind of cognitive stasis for GW attitudes. This
is reminiscent of Lord, Ross, and Lepper (1979), in which
people with a strong position tended to polarize further after
receiving (not particularly factual) information contrary to
their views. Similarly, McCright and Dunlap (2011)
highlight data indicating that climate-relevant effects of
“education level” are moderated by conservatism or party.
(Conservative or “Republican” GW denial was slightly
positively related, if at all, with education.) This (also
correlational) evidence, they claim, disproves a naïve
“knowledge deficit” view––the view that more education
can shift the public’s beliefs toward the scientific consensus
about climate change. However, their own work shows that
liberals and conservatives tend to obtain different kinds of
information. This split leaves open the possibility that wellconstructed interventions may well induce conservatives to
accept the scientific consensus (with little challenge to their
core values). Indeed, Lewandowsky, Gignac, and Vaughan
(2013) show that offering climate scientists’ consensus
boosts anthropogenic climate change acceptance.
Our laboratory has provided arguments and many
experimental findings that run counter to these
“polarization” and cognitive stasis views: For instance, even
a small amount of true information can quickly act as a
cognitive “lever” to enhance one’s understanding and
perspective on climate change (Ranney et al., 2012a)––and
many other social issues (e.g., abortion and immigration)––
and even using just a single number/statistic (Garcia de
Osuna, Ranney, & Nelson, 2004; Munnich et al., 2003;
Ranney et al., 2008). Below, we offer further experimental
results that counter the stasis view for climate education.
Notably, we analyze the full spectrum of participants, rather
than filtering for those who are already relatively extreme.
Note that new knowledge often facilitates societal shifts
and that science “education” has historically driven major
social changes—from heliocentrism replacing church
doctrine to the acceptance of a tobacco-cancer link in spite
of industry obfuscation. (We offer more such germane
evidence below.) These data-driven shifts demonstrate how
sociologists and social psychologists who hold the stasis
view must be incorrect or overly pessimistic. Whether or not
they realize it, theorists are haggling over speed, and some
2070
nations learn (e.g., to accept evolution or climate change;
Ranney, 2012) faster than others. Of course, learning or
acting too slowly can exacerbate existing problems.
We partially agree, though, with those who critique a
“knowledge deficit” view of public attitudes (cf. Dickson,
2005). Arbitrary or propaganda-like information need not
drive one toward a more empirically supported view. We
see the problem, rather, as a wisdom deficit, for which
cognitively sophisticated educators can provide the tools
that help the public better evaluate the evidence and make
choices that match their values. (See Lewandowsky et al.,
2012, for a fine discussion of such tools, particularly the
correction of misinformation.) We believe that the findings
described here will demonstrate that a well considered
educational approach is critical for public engagement.
toward climate change (as RTMD hypothesizes, and, e.g.,
Ranney, 2012 found). These findings suggest that the effects
of well-chosen aspects of education are both significant and
somewhat independent of political affiliation. Indeed,
though not reported previously, Ranney et al.’s (2012a) data
also showed that evolution acceptance was a significant
predictor of climate change acceptance even in a model
2
including the two major political parties (𝜒 (4)=12.3,
p<0.02; N.B., including other parties dramatically reduces
quality of fit for any model, likely due to small bin sizes).
Efficacy in Learning Climate Change’s Mechanism
The GW Mechanism: Extra Greenhouse Effect
Much of our laboratory’s prior research has sought to foster
worthwhile, notable conceptual changes with short activities
that involve estimations, predictions, or explanations. These
activities are followed by small amounts of feedback:
Numerically Driven Inferencing studies (NDI; e.g., Ranney
et al., 2008) have provided numeric feedback. RTMD
(Reinforced Theistic Manifest Destiny) theory, which
examines why people in the U.S. are less likely to accept
evolution and climate change than are people in peer nations
(e.g., Ranney, 2012), has yielded mechanistic interventions.
Ranney et al. (2012a) found that almost no Americans
seem to understand the basic mechanism of global warming.
Ranney et al. (2012b) includes a 400-word explanation (and
experimental stimulus) of the physical-chemical mechanism
of the greenhouse effect, here summarized: (1) Earth’s
surface absorbs (mostly visible) sunlight and then emits
infrared light, which greenhouse gases absorb, causing heat
energy to leave the atmosphere more slowly than it arrived;
(2) as people add more greenhouse gases, the Earth
experiences climate change (an added greenhouse effect). In
one survey, not a single person out of 270 (mostly public
park visitors) could correctly describe (1) and (2). Virtually
none of those surveyed could explain a key conceptual
piece: the asymmetry of how energy can reach Earth yet
then get “trapped” after it arrives (like a “leaky one-wayvalve”), due to the visible-to-infrared energy conversion.
Mechanism Knowledge is Related to GW Attitudes
Ranney et al. (2012a) found that the correlation between
mechanistic climate change knowledge and attitude toward
climate change was robust even when taking into account
political party. Mechanistic knowledge correlates with
acceptance that global warming is occurring (r=0.22,
p=0.0002) and is anthropogenic (r=0.17, p=0.005).
Anthropogenic climate change acceptance also predicted
financial “willingness to sacrifice” (χ2(4) > 32, p<0.001 for
each of four items), and one’s knowledge score predicted
two of these items (χ2(1) > 3.8, p<0.05 for both). Further,
acceptance of biological evolution (another controversial
science topic) was found to predict beliefs and attitudes
Ranney, et al. (2012a & 2012b) also provided two divergent
undergraduate samples (from UT-Brownsville, a “HispanicServing Institution,” and UC-Berkeley) with the
aforementioned 400-word description. Strikingly, this threeminute intervention roughly tripled their mechanistic
knowledge on the assessment metric. The intervention also
caused both Texas and California undergraduates to increase
their climate change acceptance. Contrasting with others’
studies noted above, our intervention focused on a
fundamental, well-researched knowledge gap, and our
assessment focused on acceptance/belief. Such contrasts
may explain the difference between observing instructional
benefits (as we have) or polarization (as others occasionally
have; cf. Lundmark, 2007). We provide further evidence
below that such interventions are applicable across broader
settings, time-frames, and populations, and that global
warming understandings and attitudes are far from static.
Study 1: A Web-based and Longevity Extension
Given the replicated demonstrations of significant attitude
changes described above, we proceeded to assess whether
the mechanism-explanation effects we had obtained were
durable or transient. This study extended prior work by
delaying the post-test several days. We were also concerned
that an “experimental demand” from the classroom setting
might have driven our prior results, so we provided the
intervention on-line; that is, we assessed whether our
materials would elicit significant attitude change even
though students participated via their own computers,
without experimenter observation. Thus we concurrently
explored both the longevity (via delay) and format (on-line)
aspects of our phenomenon. We also extended our prompts
to incorporate more demographic and introspective queries.
Methods. The instructional materials were those reported in
Ranney et al. (2012a & 2012b; the latter includes the full
400-word text of our intervention). The empirical
differences were that (a) the study was conducted online, via
the Qualtrics Inc. (Provo, UT) system, (b) eight items were
added to pre- and post-test attitude surveys to add reliability
to the related RTMD metrics (specifically, national and
religious affinities; these metrics will be reported
elsewhere), and (c) five further items were introduced
immediately following the instructional material to elicit
introspection (about embarrassment, disagreement, etc.).
2071
Undergraduates (N=80) were recruited via the Research
Participation Program (RPP), of the University of
California, Berkeley (UCB) psychology department, which
allowed us to administer a pre-test to about half of the
students (38) between eight and 26 days (𝜇=18.5 days) prior
to the study, which may have allayed test-retest effects
(although Ranney et al., 2012a, found little evidence for
them). Thus, as with Ranney et al. (2012a), some students
received the full survey testing “sandwich” while others had
no pre-test. A delayed post-test was given to all participants
between one and eight days later (𝜇=4 days); this range was
used to assess the timecourse of retention in planning
subsequent studies. We lack the power to test forgetting
over time here (although numerically, we did not find any!).
Results and Discussion. In general and as anticipated, we
replicated Ranney et al.’s (2012a) results and extended them
by finding that gains were retained over the mean, four-day,
delay. (Note: all of this piece’s measures use 1-to-9 Likert
scales.) Scored knowledge was again linked to self-rated
knowledge (r=0.5, p<0.0001) and was similar to that of
prior UCB students. Scored knowledge soared from 3.8
(pre-test) to 6.5 (post-test) and 6.3 (delayed post-test); gains
were significant (p’s<0.0001, simultaneous comparisons).
Stated GW beliefs followed a similar pattern. Mean ratings
rose from 6.20 (pre-test) to 6.54 (post-test) and were mostly
retained at 6.44 (delayed post-test)—notable gains (again)
for a 400-word text (t(79)=2.5, p=0.006 for immediate, and
t(79)=1.7, p=0.05 for delayed). The largest post-test gains
were found in agreeing with “Human activities are largely
responsible for the climate change…” (a 0.25 gain) and
certainty that global warming is occurring (a 0.19 gain). The
self-rated knowledge mean similarly increased markedly
from pre- to post-test (4.5 to 5.6). This gain’s retention,
gratifyingly, was also noted on the delayed post-test (5.2;
simultaneous comparisons for both the gain and retained
gain were again significant; p’s<0.0001). The immediate
increase in self-rated knowledge replicates results from
Ranney et al. (2012a; the results were not reported then).
In sum, Study 1 extends the finding that well-considered
information, even received online, increases anthropogenic
GW acceptance and behaviorally relevant attitudes. Further,
the conceptual changes that result from reading even 400
words have notable longevity. These effects have been
replicated with the general public as well (unpublished
data). Computer-based interventions often scale well,
enhance reliability, and prove cost-effective; so, given our
results, we recommend the online distribution of
mechanistic explanations, especially about climate change.
Altering Beliefs with Factual Numbers
The aforementioned NDI paradigm has yielded marked
attitudinal and conceptual shifts with quite minimalist
interventions. NDI and one of its procedures, EPIC (both
introduced by Ranney and students), represent a particularly
compact, well-specified intervention. EPIC participants (1)
provide an Estimate for each policy-relevant item’s
quantity, (2) state a preferred target (or monetary allocation)
Policy (or Preference) for each quantity, (3) receive true
feedback
quantities
to
Incorporate
(as
new
“Information”), and (4) indicate whether their policies have
Changed due to the feedback. With just a single wellselected quantity, the EPIC procedure’s feedback often
shifts one’s attitudes. EPIC-spawned conceptual changes are
often remarkably durable for such a small intervention (e.g.,
Ranney et al., 2008), as evidenced by increased estimation
accuracy 12 weeks after the procedure (Munnich, Ranney,
& Bachman, 2005). Therefore, we sought to employ NDI
interventions in addition to the mechanism intervention
from Study 1 and prior studies. Specifically, we presented
different participant groups with numerical information that
is relevant to global climate change acceptance. We used
numbers that were likely to boost acceptance (Studies 3 &
4), as well as numbers that we thought might erode
individuals’ acceptance of climate change (Study 2).
Study 2: Eroding Beliefs with “Evil” True Numbers
Some organizations publicize out-of-context facts to try to
undercut the reality or gravity of human-caused GW. These
are usually blatantly cherry picked, such as that Earth
slightly cooled by 0.2oF (.04% re: absolute-zero terms) from
1940 to 1975 (Jastrow, Nierenberg, & Seitz, 1991). While
surprising, this fact hardly contradicts the ever more obvious
warming trend over the last 125+ years: one can easily pick
endpoints that are oddly high or low in a noisy time series.
(The slight decrease is also explained by a planetary motion
trend.) Given this rather clear intent to mislead (Oreskes &
Conway, 2010), we (partly tongue-in-cheek) label these
numbers “evil.” Thus, Study 2’s hypotheses are that a few
misleading facts can reduce one’s (1) climate change
acceptance, (2) ratings of knowledge of the issue, and (3)
climate-change funding preferences. Of course, lest we
erode participants’ acceptance of anthropogenic climate
change more than fleetingly, we debriefed them right
afterward with more complete information––including the
mechanism and a large dose of (non-evil) relevant facts.
Methods. The survey and instructional materials were
analogous to those used in Ranney et al.’s (2012a) paperand-pencil second study. The main difference was that the
mechanism was replaced with one of two interventions. For
one version, part (n=59) of a UCB college class (N=104)
estimated each of eight items before receiving the feedback
values, with an emphasis on maximizing the quantity of
feedback numbers given to the student. To this end, this
eight-item survey included only a post-test (i.e., no pre-test),
and lacked a policy component (thus, it was an EI
intervention, lacking “P” or “C”). A more comprehensive
engagement containing only two items was administered to
the rest of the class (n=45), and this version included a pretest and extra questions about each item; we asked these
students about their surprise level after each feedback value
and requested both their climate-change funding Policies
2072
and post-feedback policy Changes regarding/versus various
UN (e.g., UNDP millennium) goals.
beliefs was administered both immediately after our
intervention and after the 18-day retention interval.
Results and Discussion. As predicted, preferences for
2
funding GW-related UN goals dropped (𝜒 (1)=22, p<0.01)
versus all eight non-climate UNDP funding alternatives.
(Unfortunately for GW as a social priority, the highest mean
pre-test preference for funding climate change initiatives
reached only a 50-50 split of available funds.) Also as
predicted, climate change acceptance significantly dropped,
from a 6.5 pre-test mean to a 6.2 post-test mean for the twoitem group (5.5% of the available room to drop on the 9point scale, t(42)=-4.3, p<0.001), and significantly to 5.9 for
the eight-item group (11% of the available room,
t(88.6)=‑2.61, p<0.005); note that these shifts were also in
the direction of ambivalence (a “5” rating), and may reflect
confusion rather than disagreement. Our third hypothesis
was also supported: self-rated knowledge fell from a mean
of 5.0 on the pre-test to 4.5 for the two-item group (12% of
the available room, t(44)=-2.5, p<0.01), and plummeted to
2.9 on eight-item survey (t(87.2)=- 5.3, p<0.001). This latter
drop of 2.1 is 53% of the available room to drop on a 9point scale, which is exceptionally large.
It is clear that even relatively educated members of the
public (e.g., undergraduates at a top-tier university) are
highly susceptible to misleading, cherry picked facts. Such
facts are clearly known to organizations attempting to
undermine the overwhelming scientific consensus about
climate change. Thus, it seems incumbent upon cognitive
science to counter the increasing sophistication with which
such organizations distribute misleading information.
Results and Discussion. Attitudes, acceptance, and beliefs
about climate change were stable after this intervention with
“saintly” numbers (pre-test: 6.71; post-test: 6.67). This
stability was unexpected (but see below for explanations),
especially because these items (as with the “evil items)
were, as anticipated, able to significantly erode self-rated
knowledge (5.3 to 4.0, t(29)=-3.6, p<0.01). This erosion was
comparable to that found with the “evil” numbers. These
items were also relatively high regarding participant
surprise, compared to Study 1’s 400-word intervention. The
mean surprise rating across Study 3’s items was 4.8, while
the mean surprise rating for the 400 words was 2.9. (All
ratings above “1” indicate some level of surprise.)
One of the most surprising numbers (a 5.2 mean) was the
near-100% of active researchers who support anthropogenic
climate change’s tenets, reflective of the strong relationship
between the perceived scientific consensus and climate
change acceptance, as Lewandowsky, Gignac, and Vaughan
(2013) report. The two numbers most similar to the statistics
in the 400 words were similarly surprising, with the rises in
atmospheric methane (+151%) and CO2 (+40%) yielding
respective surprisingness means of 5.9 and 5.1.
Despite these powerful impacts, Study 3 yielded no effect
on beliefs or attitudes. This lack of effect is counter to prior
NDI studies, in which individuals’ preferences and beliefs
were often markedly shifted by even a single number. (An
experimental silver lining here is the finding that
participants will not report greater climate change
acceptance by mere dint of experimenter demand!) One
possible explanation regards an unintended method change:
Participants in prior NDI and RTMD studies were usually
told the particular scientific/literature source—both for each
statistic that was sought and each true value provided as
feedback. Study 3 omitted this, so is possible that
participants were less compelled by the authority of this
study’s statistics, compared to those in Study 2. Another
possibility is that these UCB students were near ceiling for
acceptance, with a reluctance to admit the disturbing effects
of GW. Further, it may be that, as in Study 2, participants
were left feeling less knowledgeable—weakening any boost
these surprising numbers could have had on climate change
acceptance. Finally, perhaps students lacked an appropriate
context for integrating this information. The next study
illustrates one way to contextualize such feedback statistics.
Study 3: “Saintly” Numbers Supporting GW
Given Study 2’s observed efficacy for “evil” numbers and
the NDI paradigm’s prior successes, this study assessed the
utility of numbers that support claims of global warming.
Again partly tongue-in-cheek, we call these “saintly”
numbers. Given prior NDI studies of similarly “shocking”
magnitudes (e.g., Garcia de Osuna, et al., 2004), our
hypothesis was that the accurate feedback would increase
participants’ climate change acceptance, but diminish selfconfidence in their knowledge of the issue.
Methods. Like Study 1, Study 3 was both online and used a
UCB-RPP pre-test survey (for a subset of 30 students);
however, we increased the delay between pre-test and
intervention to a mean of 18 days. We queried the
individuals (N=60) about eight quantities. The eight items
also included questions directed at participants’ surprise and
their reactions to each number. (Monetary preferences were
left out of this version because we already observed attitude
shifts in the simplified eight-item “evil” intervention.) An
added feature of the on-screen intervention is that we could
more saliently remind individuals of each of their estimates
on the same page on which they incorporated numerical
feedback, better ensuring that they contrasted the two. As
with Study 1’s online survey, a post-test about attitudes and
Study 4: Consolidating Knowledge-Gain Effects
Study 4 explores combining (a) the replicated effect of
explaining global warming’s mechanism and (b) the
promising effect of offering representative statistics that
support understanding of GW’s effects and dangers.
Participants were 63 urban San Francisco Bay Area highschool students, who likely better represent the general
public than do the prior studies’ university undergraduates.
2073
Methods. The students were in three junior-level chemistry
classes. Much of the school (40%) is on free/reduced lunch
(a low-income marker), 95.1% is “non-white,” and just 35%
lists English as their primary language. We presented them
with (1) a more elaborated mechanistic explanation/minicurriculum, and (2) six key GW statistics. A control group
received (1) from above, with (3) six unrelated statistics.
We predicted that (i) the mechanistic explanation would yet
again yield gains in climate-change understanding and more
pro-environmental attitudes, (ii) the key statistics would
enhance such effects, and (iii) the effects would remain a
month later. Felipe (2012) describes the intervention and
results more completely. Everyone received 15 minutes’
instruction on climate change’s mechanism for each of three
days. After estimating each of the six critical climate change
quantities, the experimental group (n=33) received the true
values as feedback. The control group (n=30) received six
estimation-feedback values that proved equally surprising
but which were unrelated to climate (drawn from Ranney et
al., 2008). Each student filled out a pre-test, a post-test
(three days later; N=63), and a delayed post-test (34 days
later; N=59). Of each test’s many measures, we focus here
on scientific mechanistic knowledge, attitudes toward global
warming, and Environmental Behavioral Intentions (EBI).
We report below only the gist of some of Felipe’s (2012)
notable findings, yet provide some newer findings about the
effects of relevant, surprising numbers on the retention of
gains from the mechanistic knowledge curriculum.
Results and Discussion. Pre-test mechanism knowledge
was so close to zero that the curriculum hugely increased
both groups’ GW-mechanism understandings—by 1,619%,
on average (combined t(62)=9.31, p<0.0001). Gains in both
groups’ mean EBI scores were also quite notable
(t(62)=5.91, p<0.0001); the effects emphatically replicate
our prior three studies’ findings that show the importance of
mechanistic information in enhancing a person’s GW
understanding and “pro-environment” attitudes. Even more
importantly, the gains were significant 34 days later for both
groups (control t(27)=3.01, experimental t(28)=5.2, both
p’s<0.002), which seems an especially impressive effect for
less than one day's class out of about 170 instructional days.
(One might imagine the curriculum’s effect if it were
extended or reinforced multiple times—or given to the
general public.) Interestingly, although the control group
greatly gained by learning the mechanism, the experimental
group’s retention of their mechanistic knowledge was
significantly greater than—about twice—that of the control
group (t(48.7)=-2.61, p=0.01; planned comparison after a
significant ANOVA interaction term), suggesting that the
experimental group’s critical statistical information helped
reinforce and secure the mechanistic information. Thus, the
numbers may have “primed the pump” for learners to more
durably encode their new GW mechanism knowledge. The
differences show separate utilities for both the mechanistic
information and the statistically pertinent information––and
the suitability of our brief intervention for high school
students, and, likely, the wider public.
Students’ acceptance of climate change and concern about
its effects were both near ceiling on the pre-test (8.3 and 8.1,
respectively, thus range-restricted on the 1-9 scales); even
so, the experimental group exhibited a significant gain
(t(32)=1.76, p<0.05). Interestingly, the curriculum
inadvertently slightly increased students’ acceptance that
climate change is “just part of a natural cycle” (but to only a
modest 3.7 on the 9-point scale) rather than anthropogenic
(which slightly dropped, yet remained at about 7 on the 9point scale). Upon analysis, this modest, counterproductive
result was due to the curriculum focusing on how, prior to
humans’ influences, Earth’s evolution already yielded
greenhouse gases and a greenhouse effect. The curriculum
failed to communicate carefully that humans have caused an
extra greenhouse effect, which represents anthropogenic
climate change. This finding highlights the importance of
understanding that, while the greenhouse effect in which our
species evolved had long kept Earth from being a virtual
snowball, the extra greenhouse gases that humans have
pumped into the atmosphere during the past 250+ years are
disturbing a narrow thermal balance. Educators should
emphasize that, as with drinking ten gallons of water, a
beneficial substance can turn deadly in excess.
General Discussion
Our studies have provided an evidential medley that
effectively disconfirms the idea that GW-relevant
knowledge and attitudes are locked in cognitive stasis.
Contrary to those who over-problematize a “knowledge
deficit” (or “information deficit”) view of global warming
communication, we see a "wisdom deficit." Here (and in
Ranney et al., 2012a) we have markedly un-problematized
any deficit with our interventions’ "cognitive levers.” In
contrast, it is unlikely that offering an ill-structured list of
uncompelling facts to an unprepared mind (or thinly veiled
rhetoric; cf. Lord, Ross, & Lepper, 1979) will notably alter
beliefs or behaviors––especially for the difficult topic of
climate change. Rather, one must be sensitive to specific
(mis)understandings that may be relevant to a learner
grappling with a domain. Ultimately, we will likely need to
engage virtually all people, aiding them in connecting their
long-term values to the long-term effects of their behaviors.
Disturbingly, Study 2 showed that climate change
acceptance can be readily eroded by misleading, cherrypicked data. To guard against such “evil” misinformation,
people need the context to recognize them as the clever
propaganda that they are. Such prophylactic interventions
may represent promising targets for further educational
research initiatives (cf. Lewandowsky et al., 2012).
We are currently studying ways to disseminate the
information that we have found to elicit worthwhile
cognitive and belief changes. For instance, we are
developing on-line instructional materials (e.g., videos) that
can widely convey both global warming’s mechanism and
the statistics that reflect the scientific consensus of climate
change—so the public can more fully join that consensus.
Even if people forget an offered statistic or a mechanistic
2074
aspect, we hope they will recall being rationally convinced
of climate change’s reality, danger, and need for action.
We have shown above that on-line survey interventions,
brief curricula, and classroom lessons can have a marked
and persistent effect on one’s knowledge, understanding,
beliefs, and attitudes about global warming. Despite
arguments to the contrary, some simple cognitivelyinformed interventions may well be fundamental in building
humanity’s resolve to tackle global climate change.
Acknowledgments
We thank Daniel Reinholz, Lloyd Goldwasser, Baljeet Rai,
Myles Crain, Sarah Cohen, R. Farjadi, R. McGuane, S. Hu,
D. Kazmi, L. Nevo, K. Le, T. Charoenying, A. Galpern, B.
Balukjian, the rest of UCB’s Reasoning Group, and Jeff
Wilson. This research was supported by UCB’s Committee
on Research and by an Institute of Education Sciences (IES)
pre-doctoral training grant (# R305B090026) to UCB.
References
Barnosky, A. D. (2009). Heatstroke: Nature in an age of
global warming (2nd ed.). Washington, DC: Island Press.
Dickson, D. (2005). The case for a “deficit model” of
science communication. SciDev.net. Retrieved January
19, 2013, from http://www.scidev.net/en/editorials/thecase-for-a-deficit-model-of-science-communic.html
Felipe, J. (2012). Numerical reasoning, knowledge, and
environmental behavior regarding climate change.
Masters project, Graduate School of Education,
University of California, Berkeley.
Garcia de Osuna, J., Ranney, M., & Nelson, J. (2004).
Qualitative and quantitative effects of surprise:
(Mis)estimates,
rationales,
and
feedback-induced
preference changes while considering abortion. In K.
Forbus, D. Gentner, & T. Regier (Eds.), Proceedings of
the Twenty-Sixth Annual Conference of the Cognitive
Science Society (pp. 422-427). Mahwah, NJ: Erlbaum.
Hoffman, A. (2011). The growing climate divide. Nature
Climate Change, 1, 195–196.
Jastrow, R., Nierenberg, W., & Seitz, F. (1991). Global
warming: What does the science tell us? Energy, 16(1112), 1331–1345. doi:10.1016/0360-5442(91)90006-8
Kahan, D. M., Peters, E., Wittlin, M., Slovic, P., Ouellette,
L. L., Braman, D., & Mandel, G. (2012). The polarizing
impact of science literacy and numeracy on perceived
climate change risks. Nature Climate Change, 2, 732–735.
Lewandowsky, S., Ecker, U. K. H., Seifert, C. M., Schwarz,
N., & Cook, J. (2012). Misinformation and its correction:
Continued influence and successful debiasing.
Psychological Science in the Public Interest, 13, 106-131.
Lewandowsky, S., Gignac, G., & Vaughan, S. (2013). The
pivotal role of perceived scientific consensus in acceptance
of science. Nature Climate Change, 3, 399-404.
Lord C. G., Ross L., Lepper M. R. (1979). Biased
assimilation and attitude polarization: The effects of prior
theories on subsequently considered evidence. Journal of
Personality and Social Psychology, 37, 2098–2109.
Lundmark, C. (2007). The new ecological paradigm
revisited: anchoring the NEP scale in environmental
ethics. Environmental Education Research, 13, 329–347.
McCright, A. M., & Dunlap, R. E. (2011). The politicization
of climate change and polarization in the American
public’s views of global warming, 2001–2010. The
Sociological Quarterly, 52(2), 155–194.
Munnich, E., Ranney, M., & Bachman, M. L. N. (2005).
The longevities of policy-shifts and memories due to
single feedback numbers. In B.G. Bara, L. Barsalou, & M.
Bucciarelli (Eds.) Proceedings of the Twenty-seventh
Annual Conference of the Cognitive Science Society (pp.
1553-1558). Mahwah, NJ: Erlbaum.
Munnich, E., Ranney, M., Nelson, J., Garcia de Osuna, J., &
Brazil, N. (2003). Policy shift through NumericallyDriven Inferencing: An EPIC experiment about when
base rates matter. In R. Alterman & D. Kirsh (Eds.),
Proceedings of the Twenty-Fifth Annual Conference of
the Cognitive Science Society (pp. 834-839). Mahwah,
NJ: Erlbaum.
Oreskes, N., & Conway, E. M. (2010). Merchants of doubt:
How a handful of scientists obscured the truth on issues
from tobacco smoke to global warming. New York:
Bloomsbury Publishing.
Ranney, M.A. (2012). Why don't Americans accept
evolution as much as people in peer nations do? A theory
(Reinforced Theistic Manifest Destiny) and some
pertinent evidence. In K. S. Rosengren, S. Brem, E. Evans,
& G. M. Sinatra (Eds.) Evolution challenges: Integrating
research and practice in teaching and learning about
evolution (pp. 233-269). Oxford: Oxford University Press.
Ranney, M.A., Clark, D., Reinholz, D., & Cohen, S.
(2012a). Changing global warming beliefs with scientific
information: Knowledge, attitudes, and RTMD
(Reinforced Theistic Manifest Destiny theory). In N.
Miyake, D. Peebles, & R. P. Cooper (Eds.), Proceedings
of the 34th Annual Meeting of the Cognitive Science Society
(pp. 2228-2233). Austin, TX: Cognitive Science Society.
Ranney, M.A., Clark, D., Reinholz, D., & Cohen, S.
(2012b). Improving Americans' modest global warming
knowledge in the light of RTMD (Reinforced Theistic
Manifest Destiny) theory. In J. van Aalst, K. Thompson,
M. M. Jacobson, & P. Reimann (Eds.), The Future of
Learning: Proceedings of the Tenth International
Conference of the Learning Sciences (pp. 2-481 to 2-482).
International Society of the Learning Sciences, Inc.
Ranney, M. A., Rinne, L. F., Yarnall, L., Munnich, E.,
Miratirx, L., & Schank, P. (2008). Designing and
assessing numeracy training for journalists: Toward
improving quantitative reasoning among media
consumers. In P A. Kirschner, F. Prins, V. Jonker, & G.
Kanselaar (Eds.), International Perspectives in the
Learning Sciences: Proceedings of the 8th International
Conference for the Learning Sciences (pp. 2-246 to 2253). International Society of the Learning Sciences, Inc.
World Bank (2012). Turn down the heat: Why a 4oC warmer
world must be avoided. Washington, DC: The World Bank.
2075